PhD Thesis Marine Weather Routing– Optimization Algorithm

[Quantitative Life Sciences]

_
PhD Thesis - Marine Weather Routing– Optimization Algorithm_

Marine Weather Intelligence (MWI) is a startup whose mission is to 
contribute to the decarbonization of maritime transport and to the 
safety of ships at sea. To achieve this, MWI is developing a SaaS 
platform (web application) that will primarily integrate a solution for 
routing ships with the contribution of artificial intelligence as well 
as an automatic alert system to ensure safety against extreme weather 
phenomena.

_Abstract Thesis_
In the current maritime context, traditional deterministic models for 
ship routing often do not take into account the uncertainties related to 
environmental and operational conditions. This thesis proposes an 
innovative approach using probabilistic models and multi-objective 
optimization techniques to improve ship routing. By leveraging advanced 
machine learning algorithms, we aim to optimize ship trajectories 
considering not only travel time but also safety, energy consumption, 
and other relevant factors.

We will explore a wide range of algorithms, including for example 
reinforcement learning and simulation-based optimization, evaluating 
their ability to handle dynamic and uncertain conditions in a 
probabilistic framework. The integration of multi-objective optimization 
allows for balancing various priorities, such as minimizing journey 
duration while maximizing safety and reducing fuel consumption.

Recent deep learning techniques, such as convolutional and recurrent 
neural networks, will be used to efficiently process extensive spatial 
and temporal data. This includes the analysis of historical time series 
and real-time weather data. Furthermore, the study of historical models 
such as ERA5 will enrich our understanding of past conditions and offer 
valuable insights for improving future predictions and routing strategies.

We hope that the application of these AI methods, combined with the 
analysis of real-time and historical data, will lead to significant 
improvements in routing decisions. This research makes an important 
contribution to the maritime world by proposing practical solutions that 
improve safety, performance, and contribute to the decarbonization of 
maritime transport.

_Candidate profile

_We are looking for a motivated and talented student holding a Master 
degree with:
• Background in statistics, applied mathematics and machine learning;
• Experience in programming, preferably in Python;
• English skills allowing scienLfic communicaLon (oral/reading/wriLng).

_Details

_This PhD project is a CIFRE thesis that could start as soon as possible 
(ideally before end 2024). The PhD student will mainly work in between 
MWI in Auray and the Laboratory of Mathematics of Atlantic Brittany (UMR 
CNRS 6205) at the Université Bretagne Sud located on the Tohannic Campus 
in Vannes [link]. The student will enjoy an international and creative 
environment where research seminars and reading groups take place very 
often. Moreover, a third partner will be involved in this project, the 
International Centre for Theoretical Physics (ICTP) in Trieste, Italy. 
Travels will be planned between the 3 sites.

The student will be supervised by:
• Antonio Celani  : acelani at ictp.it
• Valentin Lefranc : valentin at marine-weather.com
• François Septier  : francois.septier at univ-ubs.fr

_Salary:

_The student will receive a salary of 35,000 euros per year before taxes.




  Marine Weather Intelligence SAS – 17 rue du Danemark – 56400 AURAY, 
France contact at marine-weather.com




Erica Sarnataro
Group Secretary
Quantitative Life Sciences
The Abdus Salam International Centre for Theoretical Physics (ICTP)
Trieste, Italy
Tel. +39-040-2240623
www.ictp.it/research/qls.aspx
e-mail: qls at ictp.it